The Spanning Tree Protocol (STP) achieves a function of eliminating a Layer 2 network loop by transmitting a protocol packet of a bridge protocol data unit (BPDU) between switches and completing spanning tree computation based on information included in the BPDU. When a network fault occurs, the Spanning Tree Protocol re-computes a blocking point and enables a redundant link to recover a network service. An STP network computes only one spanning tree, and all traffic of a virtual local area network (VLAN) are forwarded according to one spanning tree, making it impossible to implement load sharing.
The MSTP standard defines the concept of a region. The MSTP divides a switching network into multiple regions. Generally, switches in a geographical scope belong to a same region. In a region, instances are defined according to different service VLAN and a spanning tree is computed for each instance. Therefore, multiple spanning trees may be formed in each region and each spanning tree is called a multiple spanning tree instance (MSTI). A VLAN and an MSTI are associated by setting a VLAN mapping table. For different MSTIs, traffic of a VLAN in an instance may be forwarded according to a computed spanning tree topology, and load sharing of spanning trees in a region may be implemented through different topologies of various spanning trees.
When all bridges of a Layer 2 network in an area need to join a same region, the prior art uses a manual configuration manner in which operation personnel configures region information of the MSTP on each switch. A manual configuration solution, however, involves an issue about a large workload, difficult deployment, and high operation and maintenance costs.